Physiological Signal Cycle Detection via Extreme Value Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for obtaining the cycle of physiological signals, such as heartbeat and respiration, are inefficient and costly due to the high computational requirements of real-time heartbeat rate recognition, especially when using cheap ARMs, which limits the cost-effectiveness and efficiency of signal processing.
Innovation Solution
A method involving a collection device to gather vibration signals, processing these signals to obtain physiological signals, and determining cycles by identifying extreme values and calculating time differences, which reduces computational power and hardware costs by using a simple extreme value recognition algorithm.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If real-time heartbeat rate recognition method using auto-correlation function is used, then measurement precision of physiological signal cycle is improved, but device complexity and computational cost increase significantly
Solution Approach 1:
The patent extracts only the essential feature needed for cycle detection - the extreme values (peaks and valleys) of the physiological signal - and discards the complex auto-correlation computation. By focusing solely on identifying these extreme points and measuring their time intervals, the system achieves accurate cycle detection with minimal computational overhead, directly resolving the contradiction between precision and complexity.
Solution Approach 2:
The patent replaces expensive, computationally intensive algorithms with simple, lightweight operations. Instead of using resource-heavy auto-correlation functions that require significant processing power, the system employs basic extreme value detection and time difference calculation that can be executed on low-cost microcontrollers, making the solution economically viable while maintaining accuracy.
2Measurement precision
If real-time heartbeat rate recognition method is used, then measurement precision is improved, but use of energy increases due to high computational requirements
Solution Approach 1:
The patent extracts only the critical information needed for cycle detection - the extreme values of the signal - and eliminates the energy-consuming auto-correlation computation. By detecting peaks and valleys directly and calculating their time intervals, the system achieves precise cycle measurement with minimal energy expenditure, suitable for battery-powered portable devices.
Solution Approach 2:
The patent replaces energy-intensive processing algorithms with simple, low-power operations. The extreme value detection and time difference calculation require minimal computational resources, enabling the system to run on low-power microcontrollers and extend battery life while maintaining accurate physiological signal cycle detection.
3Device complexity
If simple threshold shaping method is used, then device complexity is reduced, but measurement precision of physiological signal cycle deteriorates
Solution Approach 1:
The patent introduces a dynamic confirmation mechanism that adapts to the signal characteristics. Instead of using a fixed simple threshold, the system confirms extreme values by checking if they meet specific criteria (being greater than previous extremes or less than previous valleys) and verifying time duration, thereby maintaining measurement precision while keeping the device structure simple.
Solution Approach 2:
The patent replaces the mechanical threshold shaping approach with a logical decision-making system that uses conditional judgments. By substituting the simple but inaccurate threshold method with a structured extreme value confirmation process involving comparisons and time checks, the system achieves both simplicity and accuracy in cycle detection.
Data Source
AI summary
A method for obtaining a cycle of a physiological signal includes: a collection device for collecting a vibration signal of body movements; a processor for obtaining a physiological signal by processing the vibration signal; receiving a physiological signal value and a register value, comparing the physiological signal value with the register value, and reserving one of the physiological signal value and the register value; determining the physiological signal value having a corresponding time duration, reaching a given set time to be an extreme value, wherein the time duration is a time duration of the physiological signal value received is not exceeded; restarting the procedure and determining a next extreme value; obtaining the cycle of the physiological signal by processing the at least one extreme value; and displaying the cycle of the physiological signal in a display device.